This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
| Other format: | [Raw text] | |
I've already deleted the gcc build that gave me problems, so I'll try it agian sometime later.I tried the Sun linker and assembler in "/usr/ccs/bin". With them I can at least run a simple program, but now there are
problems linking source files that use "std::basic_string<wchar_t>".
Do you have a simple testcase that fails for you? Then please file a bug (via gccbug) so this can be investigated.
We are building a cross-platform app (MacOSX, Windows, Solaris, Linux, and hopefully FreeBSD) that needs 2-byte unicode characters. In "gcc 3.0.4" on Solaris this flag wasn't necessary to make -fshort-wchar work. However, for gcc 3.2 on Linux it seems to be necessary, and I assumed it was now for Solaris as well.I'm sure I'll get it going eventually. I just found out I need another configure flag "--enable-cxx-flags="-fshort-wchar". Thank you for your help.This seems terribly wrong: the gcc/g++ defaults for wchar_t match what Solaris 8 has in <iso/wchar_iso.h>, while the -fshort-wchar flag changes wchar_t to short, creating an incompatibility with system functions. Why do you think you need this? Rainer
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |